Interpretable machine learning optimization(InterOpt)for operational parameters:A case study of highly-efficient shale gas development

An algorithm named InterOpt for optimizing operational parameters is proposed based on interpretable machine learning,and is demonstrated via optimization of shale gas development.InterOpt consists of three parts:a neural network is used to construct an emulator of the actual drilling and hydraulic fracturing process in the vector space(i.e.,virtual environment);:the Sharpley value method in inter-pretable machine learning is applied to analyzing the impact of geological and operational parameters in each well(i.e.,single well feature impact analysis):and ensemble randomized maximum likelihood(EnRML)is conducted to optimize the operational parameters to comprehensively improve the efficiency of shale gas development and reduce the average cost.In the experiment,InterOpt provides different drilling and fracturing plans for each well according to its specific geological conditions,and finally achieves an average cost reduction of 9.7%for a case study with 104 wells.

Investigation of influence factors on CO_(2) flowback characteristics and optimization of flowback parameters during CO_(2) dry fracturing in tight gas reservoirs

CO_(2) dry fracturing is a promising alternative method to water fracturing in tight gas reservoirs,especially in water-scarce areas such as the Loess Plateau.The CO_(2) flowback efficiency is a critical factor that affects the final gas production effect.However,there have been few studies focusing on the flowback characteristics after CO_(2) dry fracturing.In this study,an extensive core-to-field scale study was conducted to investigate CO_(2) flowback characteristics and CH_(4) production behavior.Firstly,to investigate the impact of core properties and production conditions on CO_(2) flowback,a series of laboratory experiments at the core scale were conducted.Then,the key factors affecting the flowback were analyzed using the grey correlation method based on field data.Finally,taking the construction parameters of Well S60 as an example,a dual-permeability model was used to characterize the different seepage fields in the matrix and fracture for tight gas reservoirs.The production parameters after CO_(2) dry fracturing were then optimized.Experimental results demonstrate that CO_(2) dry fracturing is more effective than slickwater fracturing,with a 9.2%increase in CH_(4) recovery.The increase in core permeability plays a positive role in improving CH_(4) production and CO_(2) flowback.The soaking process is mainly affected by CO_(2) diffusion,and the soaking time should be controlled within 12 h.Increasing the flowback pressure gradient results in a significant increase in both CH_(4) recovery and CO_(2) flowback efficiency.While,an increase in CO_(2) injection is not conducive to CH_(4) production and CO_(2) flowback.Based on the experimental and field data,the important factors affecting flowback and production were comprehensively and effectively discussed.The results show that permeability is the most important factor,followed by porosity and effective thickness.Considering flowback efficiency and the influence of proppant reflux,the injection volume should be the minimum volume that meets the requirements for generating fractures.The soaking time should be short which is 1 day in this study,and the optimal bottom hole flowback pressure should be set at 10 MPa.This study aims to improve the understanding of CO_(2) dry fracturing in tight gas reservoirs and provide valuable insights for optimizing the process parameters.

Hydrocarbon gas huff-n-puff optimization of multiple horizontal wells with complex fracture networks in the M unconventional reservoir

The oil production of the multi-fractured horizontal wells(MFHWs) declines quickly in unconventional oil reservoirs due to the fast depletion of natural energy. Gas injection has been acknowledged as an effective method to improve oil recovery factor from unconventional oil reservoirs. Hydrocarbon gas huff-n-puff becomes preferable when the CO_(2) source is limited. However, the impact of complex fracture networks and well interference on the EOR performance of multiple MFHWs is still unclear. The optimal gas huff-n-puff parameters are significant for enhancing oil recovery. This work aims to optimize the hydrocarbon gas injection and production parameters for multiple MFHWs with complex fracture networks in unconventional oil reservoirs. Firstly, the numerical model based on unstructured grids is developed to characterize the complex fracture networks and capture the dynamic fracture features.Secondly, the PVT phase behavior simulation was carried out to provide the fluid model for numerical simulation. Thirdly, the optimal parameters for hydrocarbon gas huff-n-puff were obtained. Finally, the dominant factors of hydrocarbon gas huff-n-puff under complex fracture networks are obtained by fuzzy mathematical method. Results reveal that the current pressure of hydrocarbon gas injection can achieve miscible displacement. The optimal injection and production parameters are obtained by single-factor analysis to analyze the effect of individual parameter. Gas injection time is the dominant factor of hydrocarbon gas huff-n-puff in unconventional oil reservoirs with complex fracture networks. This work can offer engineers guidance for hydrocarbon gas huff-n-puff of multiple MFHWs considering the complex fracture networks.

Optimization of Gas-Flooding Fracturing Development in Ultra-Low Permeability Reservoirs

Ultra-low permeability reservoirs are characterized by small pore throats and poor physical properties, which areat the root of well-known problems related to injection and production. In this study, a gas injection floodingapproach is analyzed in the framework of numerical simulations. In particular, the sequence and timing of fracturechanneling and the related impact on production are considered for horizontal wells with different fracturemorphologies. Useful data and information are provided about the regulation of gas channeling and possible strategiesto delay gas channeling and optimize the gas injection volume and fracture parameters. It is shown that inorder to mitigate gas channeling and ensure high production, fracture length on the sides can be controlled andlonger fractures can be created in the middle by which full gas flooding is obtained at the fracture location in themiddle of the horizontal well. A Differential Evolution (DE) algorithm is provided by which the gas injectionvolume and the fracture parameters of gas injection flooding can be optimized. It is shown that an improvedoil recovery factor as high as 6% can be obtained.

Optimization of shale gas fracturing parameters based on artificial intelligence algorithm

Resource-rich shale gas plays a pivotal role in new energy types.The key to scientifically and efficiently developing shale gas fields is to clarify the main factors that affect the production of shale gas wells.In this paper,according to the shale gas reservoir characteristic of the Fuling marine Longmaxi Formation,a single-well geological model was established using the reservoir numerical simulation software CMG.Then,10,000 different reservoir models were randomly generated for different formation physical parameters,completion parameters,and fracturing parameters using the Monte Carlo method,and these 10,000 models were simulated numerically.The machine learning model uses a dataset of 10,000 different geological,completion,and fracturing parameters as input and 10,000 production curves as output.Multiple machine learning regression methods were used to train and test the dataset,and the optimal method(GBDT algorithm)was selected,and the accuracy R2 of the test set of the GBDT prediction model is 0.96.A fracturing parameter optimization workflow was constructed by combining a production prediction model with a particle swarm optimizer(PSO).The process can quickly optimize the fracturing parameters and predict the production for each time by targeting the cumulative gas production under different geological conditions.The optimized parameters are Fracture Spacing,Fracture Width,Intrinsic Permeability,Fracture Half-length,Langmuir Pressure,and Langmuir Volume.The initial predicted cumulative gas production was 4.59×10^(8)m^(3),which was optimized to 4.90×10^(8) m^(3).The proposed PSO-GBDT proxy model can instantly predict the production of shale gas wells with considerable accuracy,reliability,and efficiency,which is a vital tool for optimizing fracture design.This investigation provides a solid foundation for predicting the production of unconventional gas reservoirs and for parameter optimization.

Identification welding parameters using complex criteria of quality

In the present paper,the graphical method determination of optimal welding parameters of GMAW is described.Some quality complex criteria that are the ratio of weld sizes have been proposed. These criteria allow optimizing the shape of the weld metal zone,reducing the likelihood of hot cracks formation,and reducing the stress concentration of the weld at the toe. In the proposed methodology,the earlier described equations connecting the welding parameters and the weld dimensions of mild and low-alloy steels are explored. Determination of the optimal welding parameters is carried out graphically taking into account the variance of the welding parameters. An example of using the proposed method destined to determine the optimum welding parameters is provided. It is shown that its accuracy is sufficiently good for a practical application.

Multi-objective optimization of gas metal arc welding parameters and sequences for low-carbon steel (Q345D) T-joints

Q345D high-quality low-carbon steel has been extensively employed in structures with stringent weld- ing quality requirements. A multi-objective optimization of welding stress and deformation was presented to design reasonable values of gas metal arc welding parameters and sequences of Q345D T-joints. The optimized factors included continuous variables （welding current （I）, welding voltage （U） ahd welding speed （V）） and discrete variables （welding sequence （S） and welding direc- tion （D））. The concepts of the pointer and stack in Visual Basic （VB） and the interpolation method were introduced to optimize the variables. The optimization objectives included the different combina- tions of the angular distortion and transverse welding stress along the transverse and longitudinal dis- tributions. Based on the design of experiments （DOE） and the polynomial regression （PR） model, the finite element （FE） results of the T-joint were used to establish the mathematical models. The Pareto front and the compromise solutions were obtained by using a multi-objective particle swarm optimization （MOPSO） algorithm. The optimal results were validated by the corresponding results of the FE method, and the error between the FE results and the two-objective results as well as that be-tween the FE results and the three-objective optimization results were less than 17.2% and 21.5%, respectively. The influence and setting regularity of different factors were discussed according to the compromise solutions.

Productivity model for gas reservoirs with open-hole multi-fracturing horizontal wells and optimization of hydraulic fracture parameters

Multi-fractured horizontal wells are commonly employed to improve the productivity of low and ultra-low permeability gas reservoirs.However,conventional productivity models for open-hole multi-fractured horizontal wells do not consider the interferences between hydraulic fractures and the open-hole segments,resulting in significant errors in calculation results.In this article,a novel productivity prediction model for gas reservoirs with open-hole multi-fractured horizontal wells was proposed based on complex potential theories,potential superimposition,and numerical analysis.Herein,an open-hole segment between two adjacent fractures was regarded as an equivalent fracture,which was discretized as in cases of artificial fractures.The proposed model was then applied to investigate the effects of various parameters,such as the angle between the fracture and horizontal shaft,fracture quantity,fracture length,diversion capacity of fractures,horizontal well length,and inter-fracture distance,on the productivity of low permeability gas reservoirs with multi-fractured horizontal wells.Simulation results revealed that the quantity,length,and distribution of fractures had significant effects on the productivity of low permeability gas reservoirs while the effects of the diversion capacity of fractures and the angle between the fracture and horizontal shaft were negligible.Additionally,a U-shaped distribution of fracture lengths was preferential as the quantity of fractures at shaft ends was twice that in the middle area.

Integrated construction technology for natural gas gravity drive and underground gas storage

Based on the mechanisms of gravity displacement,miscibility,viscosity reduction,and imbibition in natural gas flooding,an integrated reservoir construction technology of oil displacement and underground gas storage(UGS)is proposed.This paper systemically describes the technical connotation,site selection principle and optimization process of operation parameters of the gas storage,and advantages of this technology.By making full use of the gravity displacement,miscibility,viscosity reduction,and imbibition features of natural gas flooding,the natural gas can be injected into oil reservoir to enhance oil recovery and build strategic gas storage at the same time,realizing the win-win situation of oil production and natural gas peak shaving.Compared with the gas reservoir storage,the integrated construction technology of gas storage has two profit models:increasing crude oil production and gas storage transfer fee,so it has better economic benefit.At the same time,in this kind of gas storage,gas is injected at high pressure in the initial stage of its construction,gas is injected and produced in small volume in the initial operation stage,and then in large volume in the middle and late operation stage.In this way,the gas storage wouldn’t have drastic changes in stress periodically,overcoming the shortcomings of large stress variations of gas reservoir storage during injection-production cycle due to large gas injection and production volume.The keys of this technology are site selection and evaluation of oil reservoir,and optimization of gravity displacement,displacement pressure,and gas storage operation parameters,etc.The pilot test shows that the technology has achieved initial success,which is a new idea for the rapid development of UGS construction in China.

Effects of instantaneous shut-in of high production gas well on fluid flow in tubing

As the classical transient flow model cannot simulate the water hammer effect of gas well, a transient flow mathematical model of multiphase flow gas well is established based on the mechanism of water hammer effect and the theory of multiphase flow. With this model, the transient flow of gas well can be simulated by segmenting the curved part of tubing and calculating numerical solution with the method of characteristic curve. The results show that the higher the opening coefficient of the valve when closed, the larger the peak value of the wellhead pressure, the more gentle the pressure fluctuation, and the less obvious the pressure mutation area will be. On the premise of not exceeding the maximum shut-in pressure of the tubing, adopting large opening coefficient can reduce the impact of the pressure wave. The higher the cross-section liquid holdup, the greater the pressure wave speed, and the shorter the propagation period will be. The larger the liquid holdup, the larger the variation range of pressure, and the greater the pressure will be. In actual production, the production parameters can be adjusted to get the appropriate liquid holdup, control the magnitude and range of fluctuation pressure, and reduce the impact of water hammer effect. When the valve closing time increases, the maximum fluctuating pressure value of the wellhead decreases, the time of pressure peak delays, and the pressure mutation area gradually disappears. The shorter the valve closing time, the faster the pressure wave propagates. Case simulation proves that the transient flow model of gas well can optimize the reasonable valve opening coefficient and valve closing time, reduce the harm of water hammer impact on the wellhead device and tubing, and ensure the integrity of the wellbore.

The “fracture-controlled reserves” based stimulation technology for unconventional oil and gas reservoirs

To solve the problems facing the economic development of unconventional oil and gas, a new concept and corresponding technology system of reservoir stimulation based on "fracture-controlled reserves" are put forward. The "fracture-controlled reserves" stimulation technology is to realize the three-dimensional producing and economic and efficient development of unconventional hydrocarbon resources by forming a fracture system that well matches "sweet spots" and "non-sweet spots". The technical route of the stimulation technology is "three optimizations and controls", that is, control the scope of sand body through optimizing well spacing, control the recoverable reserves through optimizing fracture system, and control the single well production reduction through optimizing energy complement method. The "fracture-controlled reserves" stimulation emphasizes the maximization of the initial stimulation coefficient, the integration of energy replenishment, stimulation and production, and prolonging the re-fracturing cycle or avoiding re-fracturing. It aims at realizing the three-dimensional full producing and efficient development of reservoir in vertical and horizontal directions and achieving the large-scale, sustainable and high profitable development of unconventional oil and gas resources. The stimulation technology was used to perform 20 pilot projects in five typical tight-oil, shale gas blocks in China. The fracturing and producing effects of tight oil improved and the commercial development for shale gas was realized.

Application of a Support Vector Machine Algorithm to the Safety Precaution Technique of Medium-low Pressure Gas Regulators

In the gas pipeline system, safe operation of a gas regulator determines the stability of the fuel gas supply, and the medium-low pressure gas regulator of the safety precaution system is not perfect at the present stage in the Bei- jing Gas Cxroup; therefore, safety precaution technique optimization has important social and economic signific- ance, In this paper, according to the running status of the medium-low pressure gas regulator in the SCADA sys- tem, a new method for gas regulator safety precaution based on the support vector machine （SVM） is presented. This method takes the gas regulator outlet pressure data as input variables of the SVM model, the fault categories and degree as output variables, which will effectively enhance the precaution accuracy as well as save significant manpower and material resources.

The Extended Grey GM(2,1,Σexp(ct))Model and Its Application in the Predictions

The conventional grey GM(2,1)model built for the fast growing time sequence generally has big errors.To improve the modeling precision,the paper improves from the following two aspects:First,the paper transforms the accumulated generating sequence of original time sequence quantitatively to make the transformed time sequence have the better adaptability to the model;second,the paper extends the conventional grey GM(2,1)model’s structure to make the extended model meet the variation law of fast growing sequence better.The extended grey model is called the GM(2,1,Σexp(ct))model.The paper offers the parameter optimization method and the solving method of time response sequence of GM(2,1,Σexp(ct))model.Using the model and methods proposed,the paper builds the GM(2,1,Σexp(ct))models for the natural gas consumption of China and Chongqing City,China,respectively.Results show that the models built have high simulation precision and prediction precision.